Planta (1983)159:454468

P l a n t a 9 Springer-Verlag 1983

The localization, metabolism and biological activity of gibberellins in maturing and germinating seeds of Pisum sativum cv. Progress No. 9 Valerie M. Sponsel Agricultural Research Council Research Group, School of Chemistry, The University, Bristol BS8 1TS, UK

Abstract. Gibberellin A2o (GA2o), GA29 and GA29-catabolite were quantified in cotyledons, embryonic axes, and testas of Pisurn s a t i v u m cv. Progress No. 9 throughout the final stages of seed maturation and during germination. Stable isotope-labelled GAs were used as internal standards in conjunction with combined gas chromatography-mass spectrometry. Gibberellin A20 and GAz9 were mainly located in the cotyledons of maturing seeds, and GAzg-catabolite was predominantly located in the testa. Stable isotope- and radio-labelled G A 2 o and G A 2 9 w e r e fed to both intact seeds developing in vivo, and to isolated seed parts cultured in vitro. The combined results of in-vivo and in-vitro feeds indicated that GA20 is metabolised to GA29 in the cotyledons, that GA29 is transported from the cotyledons to the testa, and that GA29 is metabolised to GAz9-catabolite in the testa. Although the metabolism of G A 2 o in the cotyledons and of GA29 in the testa has been shown definitively, the mobility of GA29 has not yet been demonstrated directly. During seed desiccation and germination GA29-catabolite and products arising from it are transferred from the testa into the embryo. There is no evidence of a physiological function for GA29-catabolite in germination or early seedling growth. Use ofa growth retardant indicates that seedling growth, but not germination, is dependent on de-novo GA biosynthesis. Key words: Gibberellin (localization, metabolism) (gibberellin) - Seed maturation - Testa (gibberellin).

- Pisurn

Abbreviations: GA.=gibberellin An; GC-MS=combined gas

chromatography-mass spectrometry; HPLC = high-pressure liquid chromatography; M + = molecular ion

Introduction

The identification, quantitation, biosynthesis and metabolism of gibberellins (GAs) in developing seeds of P i s u m s a t i v u m , garden pea, have been studied in detail (Frydman et al. 1974; Frydman and MacMillan 1975; Sponsel and MacMillan 1977, 1978, 1980; Durley et al. 1979; Kamiya and Graebe 1983). However, the physiological functions of GAs in developing pea seeds have yet to be established. Applied GAs have been shown to induce parthenocarpic fruit set in pea (GarciaMartinez and Carbonell 1980) and promote pod growth of deseeded fruits (Eeuwens and Schwabe 1975; Sponsel 1982). Thus GAs in pea seeds may regulate pod development. This paper examines another possible role for GAs in pea seeds, namely that GAs are sequestered at the end of maturation for use during germination of the mature seed. It describes the quantification, localization and metabolism of GAs during the final stages of pea seed maturation and in mature and germinating seeds. The biological activity of some of these GAs is presented, and their possible significance in seed germination is discussed. Material and methods Labelled GAs. Stable isotope-labelled GAs, used as internal

standards, were [3~-ZH]GAzo (0.89 atoms 2H per molecule) prepared by Beale etal. (1980), [17--~3C]GA29 (0.97atoms a3C per molecule) prepared by Kirkwood et al. (1982) and [19-180]GA29-catabolite (0.69 atoms 180 per molecule) prepared by Gaskin et al. (1981). Gibberellins, doubly labelled with both a stable and a radio-isotope, were used as substrates for metabolic work. Gibberellin A2o, obtained from GA 3 by the method of Beale et al. (1980), was converted to the nor-ketone and was doubly labelled in a Wittig reaction to give [17-13C, 17-3H2]GA2o t Prepared by Dr. C. Willis

V.M. Sponsel: Localization, metabolism and activity of gibberellins in pea seeds (containing 0.88 atoms ~3C per molecule and with a specific radio-activity of 1.27.109 Bq m M - 1); [ 1 7 - x3C, 17-3H2]GA29 (0.91 atoms 13C per molecule and 1.1.109 Bq mM -1) was prepared by Kirkwood et al. (1982). In both substrates the scrambling of the 3H label between the 15- and 17-positions is assumed by analogy (Bearder et al. 1976) but is designated [17-3H2] for convenience.

Plant material Seeds of Pisum sativum cv. Progress No. 9 (from Asmer Seeds Ltd., Leicester, UK, and Nutting and Thoday Ltd., Longstanton, Cambridge, UK) were sown singly in 3" pots in John Innes No. 1 compost. After three to four weeks seedlings were transferred, four per 8" pot, into John Innes No. 2 compost. Plants were maintained in an unheated greenhouse under natural photoperiod between April and September. Developing seeds of known ages were dissected at 0 ~ C into testa, cotyledons and embryonic axis and were stored at -15~ prior to extraction. To obtain mature seeds, approx. 40-d-old fruits were harvested and kept in open trays at room temperature until they had completely dried. The testa was chipped off mature dry seeds with a scalpel immediately before extraction. To obtain germinating seeds and seedlings, mature dry seeds were sterilized in household bleach, rinsed in sterile water and imbibed in sterile plastic Petri dishes containing 10 ml sterile water. After 3-7 d at 25 ~ C in the dark, seedlings were separated into testa, cotyledons and embryonic axis and were deep-frozen. Feed conditions. For in-vivo feeds, doubly labelled GAs were fed into intact fruits attached to the parent plants. The substrate (in 1 gl 50% aqueous ethanol per seed) was injected through the pod wall and through the testa into one cotyledon of each seed. Treated fruits were harvested immediately and at intervals for up to 12 d. Seeds were dissected into testa and embryo (cotyledons plus axis) before freezing. For in-vitro feeds, fruits were detached from the parent plant, sterilized and opened aseptically to remove the seeds. Seeds were dissected aseptically into testa and intact embryos which were cultured separately in 100-ml Erlenmeyer flasks containing 20 ml nutrient medium (Nitsch 1951) in which ferric citrate was replaced by ferric chloride. Doubly labelled GAs were added to each flask in ethanol or methanol (100 gl), and uptake of GAs was enhanced by evacuation of flasks for i min. Incubation proceeded for 3-9 d at 25 ~ C in light on an orbital shaker. Afterwards seed parts were removed from the medium, washed and deep-frozen. Media and washings were kept too.

Extraction procedure and partitioning. Seeds or seed parts were macerated in 80% aqueous methanol and internal standards if used, were added immediately after maceration. The plant material was extracted twice and pooled methanolic extracts were reduced in volume under vacuum. Aqueous extracts were adjusted to pH 8 and were partitioned three times against equal volumes of ethyl acetate to give a neutral and basic fraction (referred to as N/B ethyl acetate). Aqueous phases were adjusted to pH 3 and were partitioned four times against ethyl acetate (acidic ethyl-acetate fraction) and, on occasions, three times against water-saturated n-butanol (acidic butanol fraction). Acidic organic phases were reduced in volume under vacuum and water was added to eliminate residual acetic acid. Extracts were taken to dryness after adding toluene and were weighed. Acidic ethyl-acetate extracts were reduced in weight by dissolving in a small volume of 80% methanol and partitioning three times against petroleum ether (boiling point 60-80 ~ C). The methanolic extracts were evaporated to dryness and weighed. Extracts containing radio-labelled material were

455

sampled for radioactivity determinations throughout partitioning.

Enzyme hydrolysis. Butanol fractions were hydrolysed by one of three enzyme preparations. Pectolytic enzyme (The Boots Co., Nottingham, UK) was shaken for 15 min. in 50 m M pH 4 phosphate buffer (1 g enzyme: 5 ml buffer) and was filtered to remove it from its Kieselguhr support. The filtrate was added to dry extracts in the ratio 1 g extract: 100 ml enzyme solution. Incubation proceeded for 2 d at 37 ~ C with shaking. Alternatively, extracts in 100 mM pH 4.5 phosphate buffer were hydrolysed with a cellulase preparation (Serva Feinbiochemica, Heidelberg, FRG), using 0.5 ml buffer and 10 mg enzyme for each gram fresh weight of plant material originally extracted. A few drops of toluene were added to prevent bacterial growth and incubation proceeded for 1 d at 35 ~ C with shaking. The third enzyme tested was a pectinol "R-10 concentrate" (Rohm and Haas, Philadelphia, Pa., USA) which was added to extracts dissolved in 50 mM pH 4.5 phosphate buffer (1 g extract: 1 ml buffer: 20 mg enzyme). Incubation was for 1 d at 50 ~ C with shaking. Hydrolysates were filtered, adjusted to pH 3 and extracted four times with ethyl acetate.

Thin-layer chromatography. Extracts were strip-loaded onto plates (20 x 20 cm) of silica gel H F (B.D.H. Chemicals, Poole, Dorset, UK), 0.4 mm thickness, which had been pre-washed with water-saturated ethyl acetate, and were developed with ethyl acetate : chloroform: acetic acid (15 : 5 : 1, by vol.). A wide strip of silica corresponding to R f 0.3~0.7 was scraped from the plates, and material was eluted from the silica gel with water-saturated ethyl acetate. All traces of silica were removed by repeated centrifugation. High-pressure liquid chromatography (HPLC). Extracts for HPLC were dissolved in approx. 0.5 ml 1% acetic acid:methanol (70: 30, v/v), centrifuged and filtered through a 0.45-pro type-HA Millipore filter. They were loaded onto a C18/5 g Hypersil column (25 cm long, 0.6 cm inner diameter) (Shandon Southern Products, Runcorn, Cheshire, UK) via a Model U6K injector with a 2-ml loop (Waters Associates (Inst.), Northwich, Cheshire, UK). A linear gradient of 30-100% methanol in 1% acetic acid (Jones et al. 1980) was delivered in 25 rain by two Waters 6000A pumps controlled by a Waters 660 solvent programmer. The flow rate was 2.5 ml rain-1. Fractions were collected at 0.5 rain intervals. Radioactivity was determined in fractions by liquid scintillation counting, and adjacent radioactive fractions were pooled prior to gas chromatography (GC) and combined gas chromatography-mass spectrometry. [13C, 3H]Gibberellin Az9 was eluted in fractions 16-18, GAz9-catabolite (detected by its absorbance at 245 nm) in fractions 21 and 22, and [a3C, 3H]GAeo in fractions 3/ 33.

Derivatization, GC and GC MS. The G C - M S procedures were modified considerably during the work described in this paper. For the quantification work described initially, methylation and trimethylsilylation were conducted as described in Sponsel and MacMillan (1980). Derivatized extracts were gas chromatographed on 2% QF1, coated on Gas-Chrom Q (80-100 mesh) (Pierce & Warringer UK, Chester, Cheshire, UK) and packed in a gas column (200 cm long, 2 mm inner diameter), which was fitted in a Pye 104 gas chromatograph (Pye-Unicam, Cambridge, UK) and was coupled to an A.E.I. MS30 mass spectrometer (Kratos, Manchester, UK). Gas chromatography was conducted at 180~ programming at 3~ min -1 and with a He flow-rate of 25 ml m i n - i . Mass spectra were obtained at 24 eV with a source temperature of 210 ~ C and a separator temperature of 190 ~ C. They were recorded at 30 s per mass

456

V.M. Sponsel: Localization, metabolism and activity of gibberellins in pea seeds

decade and were processed by a V G 2035 data system (V.G. Analytical, Wythenshawe, Manchester, UK). For the metabolic work described later, methylated extracts were trimethylsilylated using N-methyl-N-trimethylsilyltrifluoroacetamide (Pierce & Warriner (U.K.) and were made up in volume with dry dichloromethane. The G C - M S was performed in one of two ways. Those samples containing > approx. 100 ng radioactive component were gas chromatographed on an SE-30 S.C.O.T. column (44 m long 0.5 m m diameter) (Scientific Glass Engineering (UK), Milton Keynes, U K ) fitted in a Pye 104 gas chromatograph and coupled to an MS30 mass spectrometer as above. Gas chromatography was performed at 180 ~ C programming at 3 ~ min -1 to 300 ~ C, and with a He pressure of 1 bar. Spectra were recorded as described above and were processed by a V G 2050 data system. Alternatively for samples containing less than 100 ng, derivatized extracts were gas chromatographed on an OV-1 vitreous silica W.C.O.T. column [GC 2 (Chromatography), Northwich, Cheshire, UK] fitted in a DANI-3800 gas chromatograph (Kontron Instruments, St. Albans, Herts., UK). Samples were injected at 30~ in the G r o b splitless mode. The injector purge was activated after 40 s. The column temperature was taken rapidly to 60 ~ C, held for 2 min, programmed at 13~ min 1 to 150~ and then 3 ~ C min -1 to 300 ~ C. The He pressure was 2 bar. The column was coupled to a V G 7050 computerised mass spectrometer, with a source temperature of 200 ~ C and an interface temperature of 250 ~ C. Mass spectra were obtained at 24 eV and were recorded at 0.7 s per decade (1.3 s per cycle). A mixture of n-alkanes (Gaskin et al. 1971) was coinjected with each sample to obtain relative retention times. Percentages of non-labelled and stable-isotope-labelled species in the recovered G A were calculated from the relative intensities of ions in the molecularion clusters of the mass spectra (Sponsel and MacMillan 1978).

Table 1. Localization of gibberellins in seeds and seedlings of Pisum sativum cv. Progress No. 9 Age (d)

R R~

1

OH

2 H 3 OH 4 OH 5 OH

3 O

H

OH OH H OH

H H OH OH

R~

6

CH2

~, ~ "CO2H H3C CO2H 10

R1

~

H

R3

FI

~H

7 OH 8 OH 9 OH

H OH H

=H =H OH

O

' ~ ,~1 HC

R1 = H'

21 a

CH2 "CO2H

C' iT'

H3 L.j

HaC 13

"CO2CH3

Cotyledons, 2 ] Embryomc axis Testa

4.8

5.2

--

1.8

2.0

7.6 0.2 0.6

0.9 0.02 6.9

7.8

2.2

24

Cotyledons, 2 Embryonic axis Testa

0.4 0.01 0.02

27

Cotyledons, 2 Embryomc axis Testa

0.2

Cotyledons, 2 Embryonic axis Testa

0.04 --

29

-

0.2

--

0.3

8.7

-

4.1 0.1 0.2

1.4 0.02 6.8

--

3.4

3.7

--

0.2

9.0

1.5

6.6

0.01

31

Cotyledons, 2 Embryonic axis Testa

36

Cotyledons, 2 Embryomc axis Testa

O.O2

Cotyledons, 2 Embryonic axis Testa

0.03 --

39

_ c

--

0.1

-

0.3

16.8

0.1

--

1.1 0.04 0.2

6.1 0.1 11.2

3b

Cotyledons, 2 Embryonic axis Testa

-

1.3 0.04 -

4.1 0.4 0.1

7

Cotyledons, 2 Embryonic axis Testa

-

0.3 -

3.5 0.2 0.1

a Days from full bloom b Days from start of imbibition c N o t detected

12

It R1 = OH

I

ON29catabolite

Maturing seeds

CH2

HaC~ CO2H R1 R2

R3

H

GA29

Germinating seeds

2 ~ ~ C~JH ' 2 ' / O

H3C R1 R2

Gibberellin (Ixg/seed component) GA2o

Chemical structures. The structures of the compounds referred

to in the text are shown below.

Seed component

-

~'~CHa 14

Results

Localization of GAs in maturing seeds and seedlings. Seeds, aged from 21 to 39 d from anthesis, and

3- and 7-d-old seedlings were separated into testa, cotyledons and embryonic axis and were extracted. [3~-2H]Gibberellin Aao , [17-13C]GA29 and [19 -180]GA29-catabolite were added to extracts as internal standards. Acidic ethyl-acetate extracts were chromatographed on thin layers, and the levels of native GA2o (structure 1), G A 2 9 (3) and GAz9-catabolite (7) in derivatized extracts were determined by GC-MS. The appropriate amount of internal standard for each G A in each extract was determined during preliminary extractions, the results of which are not shown. In later extracts the amount of internal standard added was approx. equal to the amound of endogenous GA. These results are shown in Table 1.

V.M. Sponsel: Localization, metabolism and activity of gibberellins in pea seeds

Gibberellin A 2 o (1) and G A 2 9 (3) were located predominantly in the cotyledons, with maximal values of 4.5 tag GA20 at day 21 and 7.8 lag GA29 per cotyledon pair at day 27. The levels of G A 2 o in testas and embryonic axes were too low to determine accurately with the GC-MS system available at the time. The level of G A 2 9 in testas and axes was on most occasions less than one-tenth of that in cotyledons. Gibberellin Az9-catabolite (7) was located predominantly in the testa of maturing seeds, with 16.8 lag/testa versus 6.6 lag/cotyledon pair at day 36. The summed values per total seed of each of the GAs and approximate timing of maximal levels are consistent with the previous data on the levels of GA2o and G A 2 9 (Frydman et al. 1974; Sponsel and MacMillan 1978) and of GAz9-catabolite (Sponsel and MacMillan 1980) in intact maturing seeds. After germination there was an approx, tenfold decrease in the level of GA29-catabolite in the testa (Table 1), whereas the decline in G A 2 9 and G A z 9 catabolite in the cotyledons occurred between 3 and 7 d after germination. Further results, obtained with seeds from a different seed merchant, confirmed that GA29 was located in the cotyledons of 27-d-old seeds and that GAz9-catabolite was located mainly in the testa (Table 2). However extraction of mature, dry seeds revealed that the tenfold reduction in the level of GAzg-catabolite in the testa actually occurred during seed desiccation, not during germination, i.e. 11.2 lag/testa at day 39 (Table 1) versus 1.6 lag/ testa in dry seed (Table 2). A continued decline in GAz9-catabolite in the testa during germination was partly explained by leakage, since extraction of filter paper on which the seeds were germinated shows the presence of some GAz9-catabolite (Table 2). Metabolism of[17-13C, 17-3H2]GA2o in seeds maturing in vivo. Radio-labelled and stable-isotopelabelled G A 2 0 , applied to seeds of Progress No. 9 developing in vivo, is efficiently metabolised by 2fl-hydroxylation to give GA29 (Sponsel and MacMillan 1977). To investigate the site of metabolism within the seed, [17-13C, 17-3H2]GA20 was injected into the cotyledons of 21-d-old seeds developing in vivo (5 lag substrate/seed). Some fruits were harvested immediately after feeding, and some after 2 and 5 d, and seeds were separated into testas and embryos before extraction. The amount of radioactivity recovered in methanolic extracts of pods, testas and embryos is shown in Table 3a. At day 0, more than 90% of the applied radioactivity was in the embryo. At day 2, the ra-

457

Table 2. Localization of gibberellins in developing seeds, mature, dry seeds and seedlings of Pisum sativum cv. Progress No. 9

Age

Seed component

Gibberellin (gg/seed component) GA29

GA29catabolite

13.9 0.4

1.7 8.3

Maturing seeds 27a

Embryo c Testa

Mature seeds Embryo ~ Testa

1.5 0.05

4.5 1.6

1.0 0.03 0.04

5.5 0.4 0.6

Germinating seeds 3b

Cotyledons and axis Testa Filter paper and water

" Days from full bloom b Days from start of imbibition Cotyledons and axis

Table 3. Time course feed of [17-13C, 17-3H2]GA2o to maturing pea seeds developing in vivo : 3 H data

Day0

Day2

Day5

91.0 9.8 5.4

71.0 18.2 1.7

48.7 55.3 5.8

Embryos N/B ethyl acetate Acidic ethyl acetate Residual aqueous

1.9 83.5 0.6

2.7 87.6 4.9

1.0 78.8 17.8

Testas N/B ethyl acetate Acidic ethyl acetate Residual aqueous

-

0.4 86.1 2.8

0.3 94.6 3.0

Embryos Fractions 16-18 (GA29) Fractions 21-23 (GA29-catabolite) Fractions 31-33 (GA2o)

1.7 1.8 90.9

73.4 4.6 11.3

37.8 46.8 10.2

Testas Fractions 16-18 (GA29) Fractions 21-23 (GAzg-catabolite) Fractions 31-33 (GA2o)

-

14.5 60.5 6.7

6.2 78.9 6.5

(a) Methanolic extracts" Embryos Testas Pods (b) Solvent partitionb

(c) HPLC fractionationC

a 3H in methanolic extracts expressed as a percentage of radioactivity fed b 3H in constituent fractions expressed as a percentage of radioactivity in methanolic extracts 3H in particular fractions expressed as a percentage of radioactivity in HPLC run

458

V.M. Sponsel: Localization, metabolism and activity of gibberellins in pea seeds

Table 4. Time course feed of [17-~3C,17-3H2]GA/o to maturing pea seeds developing in vivo: ~3C data Extract

Embryos

Testas

Identity

13Co

13C1

EndogExogIdentity enous enous (gg/embryo) (pg/embryo)

Day 0

GAzo

11.7

88.3

0

3.0

Day 2

GAz9

70.7

29.3

4.9

2.0

Day 5

GAz9

62.0

38.0

1.0

0.6

_

13Co

m

9OH-GAzg- 73.0 catabolite M + 460 n.d." M § 400 n.d. 9OH-GA29catabolite M § 460 M § 458 M § 400

a3C~

Endogenous (gg/testa)

Exogenous (gg/testa)

1.1 b

0.4 b

7.7 b

1.82 b

m

27.0 n.d. n.d.

80.8

19.2

n.d. n.d. n.d.

n.d. n.d. n.d.

a Not determined b Total amount of GAz9-catabolite originally present in samples as estimated from the artefacts arising from it in old samples

dioactivity in the testa had increased and by day 5 the recovered radioactivity was almost equally distributed between testa and embryo. Neither pod extracts nor the day-0 testa extract were further analysed. The distribution of radioactivity on fractionation of embryo and other testa extracts is given in Table 3 b. The bulk of radioactivity was recovered in the acidic ethyl-acetate fractions. These fractions were dried, redissolved in 80% methanol and partitioned against light petroleum. The methanol-soluble extracts were fractionated by HPLC (Table 3 c) and pooled; derivatized fractions were examined by GC-MS. The amount of each [13C, 3H]GA recovered after HPLC was calculated from 3H data shown in Table 3, and the amount of endogenous GA relative to exogenous [13C, 3H]GA was then calculated from isotope ratios in the mass spectra (Sponsel and MacMillan 1978). Results are shown in Table 4. At day 0, more than 90% of the radioactivity in the embryo extract was at the Rvo 1 for GA2o (Table 3c) and unmetabolised [13C, 3H]GA2o was identified in relevant HPLC fractions by GC-MS (Table 4). (No endogenous GA2o was present indicating that the 21 day old seeds used in this feed were equivalent to 24 day old seeds in Table 1). By day 2 substantial radioactivity in the embryo extract was at the retention volume (Rvol) for G A 2 9 (Table 3c) and the presence of [13C, 3 H ] G A 2 9 w a s confirmed by GC-MS (Table 4). By day 5 the amount of [~3C, 3 H ] G A z 9 and of endogenous GA29 in the embryo extract had declined (Table 4). In both 2- and 5-d testa extracts there was substantial radioactivity at the Rvo1 for GA29-catabolite (Table 3c). Capillary GC-MS, which was per-

formed after a delay of several months, did not reveal [13C, 3H]GA29-catabolite but 9-hydroxy [13C, 3H]GAz9-catabolite (9) and three unidentified compounds with molecular ions (M +) at 460, 458 and 400. The mass spectra were as follows: m/z 460(M § 91%), 432(15%), 419(100%), 401(17%), 387(15%), 341(15%), 311(26%), 251(33%); m/z 458(M § 100%), 426(21%), 399(66%), 385(36%), 383(30%), 339(35%), 311(23%); and m/z 400(M +, 100%), 385(18%), 341(55%), 327(30%), 251 (23%). These compounds contained ~3C, but isotope ratios could not be calculated because unenriched spectra were not available at the time. The 9-hydroxyGAz9-catabolite (9) was characterized by Gaskin et al. (1981) and was shown to be present in extracts of pea seedlings. However, an isomer of 9-hydroxyGAe9catabolite is the major component of GAz9-catabolite samples which have stood in methanol at room temperature for several weeks. Moreover, 9hydroxyGA29-catabolite and the unidentified M + 460, M § 458 and M § 400 compounds have all been observed in samples of GAz9-catabolite which have been taken through the normal extraction procedure. They are thus assumed to be artefacts arising from the GA29-catabolite. A tentative structure (13) for the M § 458 compound has been suggested by Mr. P. Gaskin (School of Chemistry, University of Bristol, UK). The recovery of [~3C, 3H]GA29-catabolite in 2- and 5-d testa extracts was calculated from 3H data (Table 3 c) , and the isotope ratios in the molecular-ion cluster of 9-hydroxy-[13C, 3H]GAz9-catabolite were used to determine the amount of endogenous GA29-catabolite in extracts (Table 4).

V.M. Sponsel: Localization, metabolism and activity of gibberellins in pea seeds

In summary, [~3C, 3H]GA2o applied to seeds maturing in vivo was efficiently 2fl-hydroxylated within the embryo to give [13C, 3H]-GAz9 in approx. 44% yield over 2 d. The metabolism probably occurs at the site of application, i.e. in the cotyledons. The [t3C, 3H]GA29 was further metabolised in the 5-d feed to [13C, 3H]GAz9-catabolite (identified in old samples as artefacts arising from it), which accumulated preferentially in the testa. Isotope ratios showed that metabolism of applied [13C, 3H]GA2o and localization of its metabolites parallelled the metabolism and distribution of endogenous GAs in the maturing seed. Table 5. Time course feed of [17-13C, 17-3H2]GA29 to maturing pea seeds developing in vivo: 3H data Day

3H in methanolic extracts as a percentage of radioactivity fed Pods

Testas

Embryos

2 4 7

7.6 5.5 5.3 6.0 3.3

4.8 21.2 34.2 45.2 53.4

94.8 80.1 71.1 55.7 46.8

Mature seeds

3.2

31.9

62.9

3-d-old seedlings

2.9

3.5

84.2

0 1

459

Metabolism 0f [1~13C, 17-3Hz]GA29 in seeds developing in vivo. [17-x3C, 17JH2]Gibberellin A 2 9 was injected into cotyledons of 28-d-old seeds developing in vivo (10 gg substrate/seed). Fruits were harvested immediately and at intervals for up to 13 d. Seeds from the last harvest were desiccated at room temperature. Half of the dry seeds were extracted, and half were imbibed for 3 d before extraction. The results of the feed are discussed in two parts. The time course from 0-7 d traces the metabolism of [13C, 3H]GA29 in maturing seeds, and from 7 d onwards it follows the fate of [13C, 3 H ] G A 2 9 and its metabolites in desiccating and germinating seeds. At day 0, nearly 95% of applied radioactivity was in the embryos (Table 5), the amount of radioactivity in the testa increased until day 7, when it exceeded that in the embryos. On fractionation of extracts, most of the radioactivity was present in acidic ethyl-acetate fractions (Table 6). Figure 1 Embryos

Testas

GA29 GA29-catob.

Bqxl0 1 o LO I

I

Not analysed contained